Canon 35mm f/2 IS vs Sigma 35mm f/1.4 Art bokeh comparison

Hi Guys,

Just in case anyone's interested, I did a quick comparison between my Canon 35mm f/2 IS and a Sigma 35mm f/1.4 Art the other day. All things considered they are both excellent pieces of glass. In real-world-non-MTF-chart-land they are both equally as sharp in my opinion and so I won't bore people with pixel peeping sharpness tests as they're pointless - they're so close. The Sigma is better at closer focusing distances wide open as the Canon ghosts a little at the edges there (if full frame sharpness is important to you at close distances wide open but there are very few instances where this is even an issue as we will usually have a subject in focus and the rest not), but beyond close distances they're very similar indeed. At some apertures (f/4 for example) the Canon is a teeny bit sharper near the centre than the Sigma (at least in these two copies) but not to the point of significance, only if pixel peeping. No lens is perfect.

Aaaaaanyway. One thing which I DID want to see is the bokeh quality between the two of them at the same apertures - or rather, to be more precise, how good the background to foreground separation (or "3D-ness") is between them as in a 35mm lens this is something which interests me.

On paper you'd assume that at f/2 (the lowest common widest aperture) the background to foreground blurry ratio doodaah would be similar, but it turns out they aren't at all.

Here's a shot I set up in a difficult scene (natural backlight light, foliage and lots of noise behind the subject) because as we all know, foliage is one of the worst scenarios for smooth bokeh so if any nervousness was going to rear its head, it would be here.

The following shots were loaded into LR4 and the only thing done to them was vignette correction via the lens profile tool. Distortion correction wasn't touched to prevent any artefacts (if any) appearing in the images. They were both metered individually for ETTR and the focus point is the same for both - the very last fret on the guitar in the centre of the image.

They were taken from approximately 1 metre away from the subject on a tripod and focused manually via Live View. The Sigma looks a teeny bit closer so I guess one of them is slightly bigger or smaller than 35mm (as the tripod and camera weren't moved) but not enough to make a significant difference to this test I'd say, although I could always repeat it and move the sigma back by 1cm or so to accommodate for the frame size difference but I feel we'd be splitting hairs.

Canon 35mm @ f/2

Sigma 35mm @ f/2

Sigma 35mm @ f/1.4 - just to see how much more of an effect the extra stop has

100% Comparison between the Canon (left) and Sigma (right) @ f/2

Canon (left) Sigma (right) @ f/2 side by side

Now, this is all subjective and down to personal preference, and this is an extreme situation for bokeh I know, but in my opinion the Sigma images just "pop" out of the frame because the background / foreground separation is so much greater. In fact you don't even need to stop down to f/1.4 to achieve this as it's lovely even at f/2 (which means hardly any vignette to correct, so less potential noise in the corners if that's an issue for you) but I've included the f/1.4 image just so you can see.

When I looked at the other photos I took in LR, the Sigma matches the Canon's blur (when at f/2) when it's at f/2.8. In fact, at f/2.8 the Sigma and the Canon are very similar it's hard to tell them apart.

Even in the final side-by-side image I've included and viewing them small and on the web, the side by side comparison illustrates this clearly to my eyes. The Sigma just has "something" to it even though the differences are quite small. It's as though your eye's peripheral vision isn't drawn to the rest of the frame when you look at the guitar, but in the Canon image, it is.

So there you go. There have been plenty of reviews of both these lenses on the net but this is one aspect which was missing for me so I hope it's useful for you so that you may draw your own conclusions about the "3D-ness" of these two lenses. Again, this isn't a review of the lenses as a whole and how the Sigma doesn't have IS and the Canon does blah blah, it's just a specific aspect of the two lenses.

Cheers,

Mark.